**4.2 Results and discussion**

The hydrogenogenic reactor (R1) operation started with a HRT of 24 h, OLR of 10 g COD L-1 d-1 and pH 5.8. Under these conditions hydrogen production was as low as 0.12 L H2 d-1 (0.027 L H2 L-1 d-1) (Fig. 13). It was noticed methane presence in biogas up to 19% v/v, while hydrogen concentration was still very low (up to 10% v/v). In order to inhibit methane production, the HRT was reduced to 12 h and OLR increased to 20 g COD L-1 d-1 on day 11. After 5 days, the HRT was increased to 24 h. Hydrogen content in biogas increased to the average value 15.7% v/v and methane was still present (<8% v/v). According to Yang et al. (2007), HRT shorter than 24 h does not favor the biohydrogen generation from cheese whey wastewater, but other researchers stated that short HRT could help to control the methanogenic reaction in hydrogenogenic phase (Castellό et al., 2009). Then, the HRT was again set at 12 h, OLR increased to 20 g COD L-1 d-1 and pH was decreased to 5.2 (day 24). It was seen that although the methanogenic bacteria was assumed to be washed out from the bioreactor under short HRT, the inhibition of methanogenic bacteria activity should be coupled with pH decrease. Liu et al. (2006) found that pH is the most critical factor for inhibition of methanogenesis and the optimum pH should be around 5.0 – 5.5. According to the literature, the optimum pH range for lactose (or whey) acidogenesis is between 6 and 6.5 (Venetsaneas et al. 2009). Antonopoulou et al. (2008) reported, that high concentration of hydrogen (over 25% v/v) in the gas phase was when the pH in the reactor was maintained at 5.2 ± 0.1. Wang et al. (2006) stated, that pH of 5.5 should be avoided in the biohydrogen fermentation process because at that level of pH, the propionic-acid type fermentation commonly occurred. The accumulation of propionic acid can lead to lower efficiency of methanogenic phase followed the hydrogenogenic phase. Mohan et al. (2007) found that pH 6 was the optimum for effective H2 yield. However, maintaining pH at 6 or above is difficult because of large amounts of fatty acids generation. At this study, the pH in the hydrogenogenic reactor was initially maintained on the level of 5.8. As the pH was reduced down to 5.2 (coupled with HRT shortening and OLR increasing), methane content in biogas faded out.

Feasibility of Bioenergy Production from

concentration in biogas.

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

mg L-1 on average.

Gas production rate (L L-1

d-1)

Ultrafiltration Whey Permeate Using the UASB Reactors 215

The effluent from R1 was collected in the storage tank from where it was pumped to the R2. The average pH at the outlet of R1 was 5.0 with a standard deviation of 0.3. According to the literature, optimum pH for methanogenic bacteria is between 6.0 and 7.5 (Mohan et al. 2007). In R2 the pH was maintained on the level of 7.2 (±0.05). The R2 was fed with the R1 effluent from 51 day of experimental period, after the R1 had reached the steady-state. The average COD concentration of the R1 effluent was 5770 mg L-1 thus the R2 reactor was operated at OLR of about 2 g COD L-1 d-1. The majority of COD was composed of VFA (total of all acids was 5147 mg L-1 on average) generated during acidogenic fermentation step in R1. Fig. 14 shows the generation of biogas and methane produced throughout the experimental period. After the first 12 days, the reactor performance was stable and the average biogas and methane production rates were 3.0 L d-1 (0.67 L L-1 d-1) and 2.2 L CH4 d-1 (0.47 L CH4 L-1 d-1), respectively. The methane content in biogas approached 71% v/v. Venetsaneas et al. (2009) found that about 1 L CH4 d-1 was achieved in two-stage process for cheese whey fermentation resulting in 68% v/v methane production which is comparable to this study. The methanogenic digester was operated at an HRT of 20 d and OLR of 2.5 g COD L-1 d-1. Liu et al. (2006) achieved better methane production from household solid waste in the two-stage fermentation process of 11.5 L CH4 d-1 with a 65% v/v methane

51 54 57 60 63 66 69 72 75 78 81 84

Biogas Methane TVFA

Time (d)

Fig. 14. Gas production and TVFA concentration in R2 throughout the experimental period

The two-stage fermentation system showed 95% of COD removal efficiency. Biohydrogen generation was connected with COD removal, thus the substrate degradation (as COD reduction) was 42.3% on average. It showed that more TVFAs were converted to biogas in the methanogenic stage. The concentration of remaining TVFAs in the effluent of R2 was 485

During the study it was noticed a partial fermentation of raw UF whey permeate, although it was maintained at 4°C before used. VFA concentration increased periodically from 55 mg L-1 to 2220 mg L-1 after thawing for 20 hours. It seems that the UF whey permeate feedstock should be supplemented with NaHCO3 to increase its alkalinity level before frozen (Castellό

0

200

400

600

TVFA (mg L-1)

800

1000

1200

Fig. 13. Hydrogen production in R1 throughout the experimental period

The experimental results demonstrated by Azbar et al. (2009) showed that OLR did not result in any statistically significant change in hydrogen production rate from cheese whey when anaerobic reactors were operated using a range of OLR of 21, 35, 45 g COD L-1 d-1 and a constant HRT of 24 h. Moreover, lower HRT values (e.g. 1 d) increased hydrogen production rate. Mohan et al. (2007) found, that increasing OLR lead to periodic decreasing in biohydrogen production rate from dairy wastewater. This was attributed to the adaptation of microbial inoculum to higher substrate loading rate which is parallel to this results (Fig. 13). High substrate loading rate shows higher availability of substrate resulting in active substrate metabolism leading to higher H2 yield (Mohan et al., 2007).

During the experiment, after reaching steady-state conditions (the standard deviations of monitored parameters were within 5%) on day 51, the average rate of hydrogen production was 0.97 L H2 d-1 (0.21 L H2 L-1 d-1) (Fig. 13), hydrogen content in biogas increased to 29,8% v/v and methane was present in a low concentration (<1% v/v). Yang et al. (2007) reported hydrogen production ranging from 0.264 to 0.312 L H2 L-1 d-1 (0.396 – 0.468 L H2 d-1) for OLR of 10 g COD L-1 d -1 and a HRT of 24 h using wastewater made from dry whey permeate powder. Biohydrogen contents in biogas fluctuated between 22 and 26% v/v. Castellό et al. (2009) obtained 0.55 L H2 d-1 (0.122 L H2 L-1 d-1) for HRT of 12 h and OLR of 20 g COD L-1 d-1 from cheese whey. Biohydrogen content in biogas ranged from 20 to 30%. Hydrogen production between 0.3 and 7.9 L H2 L-1 d-1 (2.5 L H2 L-1 d-1 on average) from dairy wastewater was reported by Azbar et al. (2009). Venetsaneas et al. (2009) operated two-stage anaerobic reactors using cheese whey as a fermentation substrate. During the hydrogenogenic stage they achieved biohydrogen production rate of 0.96 L H2 d-1 (1.92 L H2 L-1 d-1) and the percentage of hydrogen in the gas phase was 32.0±1.9% v/v (OLR 15 g COD L-1 d-1, HRT 24 h). Liu et al. (2006) achieved hydrogen production rate of 0.64 L H2 d-1 (1.6 L H2 L-1 d-1) and hydrogen content in the biogas of 42% v/v from household solid waste in the two-stage fermentation process.

H2 % H2

3 12 21 30 39 48 57 66 75 84

Fig. 13. Hydrogen production in R1 throughout the experimental period

in active substrate metabolism leading to higher H2 yield (Mohan et al., 2007).

Time (d)

The experimental results demonstrated by Azbar et al. (2009) showed that OLR did not result in any statistically significant change in hydrogen production rate from cheese whey when anaerobic reactors were operated using a range of OLR of 21, 35, 45 g COD L-1 d-1 and a constant HRT of 24 h. Moreover, lower HRT values (e.g. 1 d) increased hydrogen production rate. Mohan et al. (2007) found, that increasing OLR lead to periodic decreasing in biohydrogen production rate from dairy wastewater. This was attributed to the adaptation of microbial inoculum to higher substrate loading rate which is parallel to this results (Fig. 13). High substrate loading rate shows higher availability of substrate resulting

During the experiment, after reaching steady-state conditions (the standard deviations of monitored parameters were within 5%) on day 51, the average rate of hydrogen production was 0.97 L H2 d-1 (0.21 L H2 L-1 d-1) (Fig. 13), hydrogen content in biogas increased to 29,8% v/v and methane was present in a low concentration (<1% v/v). Yang et al. (2007) reported hydrogen production ranging from 0.264 to 0.312 L H2 L-1 d-1 (0.396 – 0.468 L H2 d-1) for OLR of 10 g COD L-1 d -1 and a HRT of 24 h using wastewater made from dry whey permeate powder. Biohydrogen contents in biogas fluctuated between 22 and 26% v/v. Castellό et al. (2009) obtained 0.55 L H2 d-1 (0.122 L H2 L-1 d-1) for HRT of 12 h and OLR of 20 g COD L-1 d-1 from cheese whey. Biohydrogen content in biogas ranged from 20 to 30%. Hydrogen production between 0.3 and 7.9 L H2 L-1 d-1 (2.5 L H2 L-1 d-1 on average) from dairy wastewater was reported by Azbar et al. (2009). Venetsaneas et al. (2009) operated two-stage anaerobic reactors using cheese whey as a fermentation substrate. During the hydrogenogenic stage they achieved biohydrogen production rate of 0.96 L H2 d-1 (1.92 L H2 L-1 d-1) and the percentage of hydrogen in the gas phase was 32.0±1.9% v/v (OLR 15 g COD L-1 d-1, HRT 24 h). Liu et al. (2006) achieved hydrogen production rate of 0.64 L H2 d-1 (1.6 L H2 L-1 d-1) and hydrogen content in the biogas of 42% v/v from household solid waste in the

0

two-stage fermentation process.

0.05

0.1

Hydrogen production rate (L L-1

d-1)

0.15

0.2

0.25

H2

content in biogas (%)

The effluent from R1 was collected in the storage tank from where it was pumped to the R2. The average pH at the outlet of R1 was 5.0 with a standard deviation of 0.3. According to the literature, optimum pH for methanogenic bacteria is between 6.0 and 7.5 (Mohan et al. 2007). In R2 the pH was maintained on the level of 7.2 (±0.05). The R2 was fed with the R1 effluent from 51 day of experimental period, after the R1 had reached the steady-state. The average COD concentration of the R1 effluent was 5770 mg L-1 thus the R2 reactor was operated at OLR of about 2 g COD L-1 d-1. The majority of COD was composed of VFA (total of all acids was 5147 mg L-1 on average) generated during acidogenic fermentation step in R1. Fig. 14 shows the generation of biogas and methane produced throughout the experimental period. After the first 12 days, the reactor performance was stable and the average biogas and methane production rates were 3.0 L d-1 (0.67 L L-1 d-1) and 2.2 L CH4 d-1 (0.47 L CH4 L-1 d-1), respectively. The methane content in biogas approached 71% v/v. Venetsaneas et al. (2009) found that about 1 L CH4 d-1 was achieved in two-stage process for cheese whey fermentation resulting in 68% v/v methane production which is comparable to this study. The methanogenic digester was operated at an HRT of 20 d and OLR of 2.5 g COD L-1 d-1. Liu et al. (2006) achieved better methane production from household solid waste in the two-stage fermentation process of 11.5 L CH4 d-1 with a 65% v/v methane concentration in biogas.

Fig. 14. Gas production and TVFA concentration in R2 throughout the experimental period

The two-stage fermentation system showed 95% of COD removal efficiency. Biohydrogen generation was connected with COD removal, thus the substrate degradation (as COD reduction) was 42.3% on average. It showed that more TVFAs were converted to biogas in the methanogenic stage. The concentration of remaining TVFAs in the effluent of R2 was 485 mg L-1 on average.

During the study it was noticed a partial fermentation of raw UF whey permeate, although it was maintained at 4°C before used. VFA concentration increased periodically from 55 mg L-1 to 2220 mg L-1 after thawing for 20 hours. It seems that the UF whey permeate feedstock should be supplemented with NaHCO3 to increase its alkalinity level before frozen (Castellό

Feasibility of Bioenergy Production from

**6. Acknowledgements** 

**7. References** 

technologies are developed to convert the waste biomass.

biogas, biohydrogen) has been successfully tested in this study.

Ministry of Science and Higher Education, Poland in 2007 – 2008.

Ministry of Science and Higher Education, Poland in 2010 – 2013.

83-60012-74-1, Warsaw, Poland

Vol.82, pp. 1-32, ISSN 0724-6145

(January 2008), pp. 110-119, ISSN 0960-8524

No.17, (September 2009), pp. 7441-7447, ISSN 0360-3199

2252-2259, ISSN 0960-8524

Ultrafiltration Whey Permeate Using the UASB Reactors 217

hydrogen produced in the world is obtained from natural gas, which is not environmental friendly. Therefore, a significant increase in biofuels production would be possible only if

Dairy industry, like most other food industries, generates strong wastewaters characterized by high COD concentrations representing their high organic content (Demirel et al., 2005). Whey is by-product of milk processing and is abundantly obtained during cheese production. According to Najafpour et al. (2008), worldwide cheese production generates more than 145 million tonnes of liquid whey per year. In the case of deproteination of whey for the production of a valuable human food additive, the residual whey permeate is still a waste with high COD and must be treated before disposal. Due to its lactose major component, whey permeate is a well defined and suitable substrate for anaerobic digestion (Kourkoutas et al., 2002; Najafpour et al., 2008; Venetsaneas et al., 2009; Zafar & Owais, 2006). UF whey permeate fermentation in UASB reactors to produce biofuels (bioethanol,

The study of bioethanol production was supported by a grant N523 049 32/1753 from

The study of biohydrogen production was supported by a grant N N 523 555138 from

Agricultural Market Agency (August 2009). Dairy Market in Poland (in Polish), ISBN 978-

Aktaş, N.; Boyac, İ.; Mutlu, M. & Tanyolaç A. (2006). Optimization of lactose utilization in

Angelidaki, I.; Ellegaard, L. & Ahring, B.K. (2003). Applications of the Anaerobic Digestion

Antonopoulou, G.; Gavala, H.N.; Skiadas, I.; Angelopoulos, K. & Lyberatos, G. (2008).

Azbar, N.; Dokgöz, F.; Keskin, T.; Korkmaz, K.S. & Syed, H.M. (2009). Continuous

Balat M. & Balat H. (2009). Recent trends in global production and utilization of bio-ethanol

Castellό, E.; Garcīa y Santos, G.; Iglesias, T.; Paolino, G.; Wenzel, J.; Borzacconi, L. &

deproteineted whey by Kluyveromyces marxianus using response surface methodology (RSM). *Bioresource Technology*, Vol.97, No.18, (December 2006), pp.

Process. Biomethanation II. *Advances in Biochemical Engineering/Biotechnology*,

Biofuels generation from sweet sorghum: Fermentative hydrogen production and anaerobic digestion of the remaining biomass. *Bioresource Technology*, Vol.99, No.1,

fermentative hydrogen production from cheese whey wastewater under thermophilic anaerobic conditions. *International Journal of Hydrogen Energy*, Vol.34,

fuel. *Applied Energy*, Vol.86, No.11, (November 2009), pp. 2273-2282, ISSN 0306-2619

Etchebehere, C. (2009). Feasibility of biohydrogen production from cheese whey using a UASB reactor: Links between microbial community and reactor

et al., 2009). It allows to reduce the lactic acid concentration in the influent, because it was reported, that the conversion efficiency of lactic acid to hydrogen is much lower than that of glucose or sucrose (Guo et al. 2010). The coexistence of LAB (lactic acid bacteria) and the hydrogen producing bacteria was investigated by Noike et al. (2002). They found, that hydrogen fermentation was replaced by lactic acid fermentation caused by LAB present in the raw wastewater. Their inhibitory effect on hydrogen production could be explained by excretion of bacteriocins (Noike et al., 2002).
